Intel
/

opt-1.3b-int4-inc-recipe

Model card Files Files and versions Community

weiweiz1 commited on Nov 6

Commit

14a5007

•

1 Parent(s): 00f1d7c

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -85,8 +85,8 @@ print(tokenizer.decode(model.generate(**tokenizer(prompt, return_tensors="pt").t
 ## Accuracy Result
-| Metric         | FP16   | INT4   |
-| -------------- | ------ | ------ |
 | Avg.           | 0.4405 | 0.4315 |
 | mmlu           | 0.2502 | 0.2510 |
 | lambada_openai | 0.5789 | 0.5389 |

 ## Accuracy Result
+  | Metric   <img width=200>     | FP16  <img width=200>   | INT4  <img width=200>   |
+| :------------------ | :---------- | :---------- |
 | Avg.           | 0.4405 | 0.4315 |
 | mmlu           | 0.2502 | 0.2510 |
 | lambada_openai | 0.5789 | 0.5389 |